Significant progress has been made in myeloma therapy, including >20% 10 year event-free survival (EFS). However, the disease remains largely incurable, and outcome in myeloma is difficult to predict. There is no reliable early surrogate marker for long-term outcome that would allow adjustment of treatment in an individual patient to avoid excessive or inadequate treatment. To further improve outcome, we must have better tools to measure the degree of tumor reduction after intensive therapy and a better understanding of the myeloma-specific genetic features related to poor or excellent survival. As currently defined, complete remission (CR) is a poor reflection of tumor load, due in part to the focal nature of myeloma and the heavy reliance of CR on myeloma (M) protein measurements, knowing that there is a major inter- and intra-patient variability in immunoglobulin production per myeloma cell. We hypothesize that predicting treatment outcome will not only depend on accurate assessment of the remaining tumor load after therapy, but also on genetic characteristics of an individual's myeloma cells. In Aim 1, we will determine the tumor burden after tandem transplants using CDR3-PCR technology in patients with high-risk myeloma (i.e., relapsing during the first 2 years after the first transplant) and in patients with low-risk myeloma (i.e., EFS >4 years). We will assess minimal residual disease (MRD) in random bone marrow and in peripheral blood as well as in focal lesions identified by MRI/PET scan. In Aim 2, we will identify the gene expression profile (GEP) associated with high risk of early progression versus prolonged remission and describe its component genes and molecular pathways. We will also construct validated prediction models. GEP will be performed at baseline, after induction therapy, and at relapse, as well as on focal MRI/PET lesions persisting in remission. This will allow us to evaluate whether a major resistant clone is already present at diagnosis, whether induction therapy acts to select out a small subclone of resistant myeloma cells, or whether resistant myeloma develops mainly as the consequence of treatment. Key genes related to poor outcome that are thus identified and validated can be targeted by specific novel therapies to maintain remissions. Distinction between the importance of absolute tumor reduction versus genetic characteristics as the major cause of treatment failure and dissecting genetic characteristics at baseline, follow-up, and relapse will provide invaluable information about the progression of myeloma. This information could then be used to apply better treatment approaches, ensuring some patients (low-risk) are not over-treated and others (high-risk) receive adequate treatment. These findings can serve as a model for treatment approaches in other chronic lymphoproliferative and myeloproliferative malignancies, which exhibit a similar disease progression. [unreadable] [unreadable] [unreadable]